You can read more about what the pipeline does here
The pipeline contains the following files/folders in rough order of importance:
main.nf
the main Nextflow script which executes all of the different processes when runnextflow.config
the main configuration file - it includes defaults for all parameters, containers & profiles/links to other configsmain.pbs
bash script to submit SLURM jobREADME.md
high-level documentation for the pipeline including links to all of the documentationdocs
more detailed documentation for the pipelinepipeline_overview.md
this overview of the pipeline structurerun_on_cloudos.md
documentation for how to run the pipeline on Lifebit's platform CloudOSrun_on_sumner.md
documentation for how to run the pipeline on JAX's HPC Sumnerusage.md
documentation for running the pipeline and parameters
conf
configuration filesexamples
contains examples data filescontainers
containsDockerfile
s (instructions) to build the Docker imagesoriginal_scripts
contains the original Bash & Nextflow scripts.gitignore
contains a list of files for git to ignore (not to commit to the repository)
The file structure looks as follows:
├── README.md
├── bin
│ ├── LU_postprocessing.R
│ ├── correct_gene_names.R
│ └── prepDE.py
├── conf
│ ├── examples
│ │ ├── MYC_MCF10A_0h_vs_MYC_MCF10A_8h.config
│ │ ├── human_test.config
│ │ └── test.config
│ └── executors
│ ├── base.config
│ ├── google.config
│ └── sumner.config
├── containers
│ ├── rmats4
│ │ ├── Dockerfile
│ │ └── environment.yml
│ └── splicing-pipelines-nf
│ ├── Dockerfile
│ ├── environment.yml
│ └── tagXSstrandedData.awk
├── docs
│ ├── github.md
│ ├── pipeline_overview.md
│ ├── run_on_cloudos.md
│ ├── run_on_sumner.md
│ └── usage.md
├── examples
│ ├── analyses
│ │ ├── MCF10_MYCER.datafiles.csv
│ │ ├── MYC_MCF10A_0h_vs_MYC_MCF10A_8h
│ │ │ ├── reads.csv
│ │ │ ├── reads_google_cloud.csv
│ │ │ └── rmats_pairs.txt
│ │ └── PRJNA453538.SraRunTable.txt
│ ├── assets
│ │ ├── TruSeq3-PE.fa
│ │ ├── TruSeq3-SE.fa
│ │ └── multiqc_config.yaml
│ └── testdata
│ ├── human_test
│ │ ├── get_human_paired_end_replicates.sh
│ │ ├── human_test_reps.csv
│ │ ├── human_test_reps_google.csv
│ │ └── rmats_pairs.txt
│ └── single_end
│ ├── rmats_pairs.txt
│ ├── test_reps.csv
│ └── test_reps_google.csv
├── main.nf
├── main.pbs
├── nextflow.config
└── original_scripts
├── bash
│ ├── pipeline_splicing_with_arguments_parallel_part1.pbs
│ ├── pipeline_splicing_with_arguments_parallel_part2.pbs
│ ├── postprocessing.pbs
│ └── run_pipeline_Olga_job1.sh
├── nextflow
│ ├── main.nf
│ └── rMATS_pipeline_samtools.nf
└── postprocessing
├── LU_postprocessing.R
├── README.md
├── config_for_rmats_and_postprocessing.txt
├── config_for_splicing_pipeline_part2.txt
├── correct_gene_names.R
├── pipeline_splicing_with_arguments_parallel_part2_withGFFcompare.pbs
├── prepDE.py
└── rmats_and_postprocessing.pbs